Company Description
Glovo is part of the Delivery Hero Group, the world's pioneering local delivery platform, our mission is to deliver an amazing experience—fast, easy, and to your door. We operate in around 65 countries worldwide. Headquartered in Berlin, Germany. Delivery Hero has been listed on the Frankfurt Stock Exchange since **** and is part of the MDAX stock market index.
Job Description
The
MarTech Data Science
team drives performance marketing efficiency through advanced measurement, optimization, and scalable modeling.
We operate in a highly complex and ambiguous environment where:
Ground truth is often unobservable (incrementality vs attribution)
Clean experimentation is not always feasible
Decisions must be made under uncertainty with imperfect data
We build solutions for budget allocation, incrementality measurement, and bidding, enabling the business to invest each euro where it generates the highest impact.
YOUR MISSION
We are looking for a
Data Scientist
to own models and features end-to-end within larger MarTech initiatives, and contribute to the team's measurement and optimization solutions. Operating at a general level, you will design solutions that serve the entire Delivery Hero portfolio, including brands like Glovo, Talabat, and PedidosYa. You will work autonomously on well-defined problems and collaborate closely with senior team members on more ambiguous ones.
This role requires solid foundations in causal inference and modeling, combined with the ability to communicate clearly with technical and non-technical stakeholders and translate marketing problems into reliable, production-grade solutions.




THE JOURNEY
Own features and components end-to-end within MarTech measurement & optimization
Take ownership of models and features inside larger initiatives, from data exploration to production deployment
Contribute to defining success metrics and modeling approaches, with senior guidance on more ambiguous problem framings
Balance methodological rigor with business constraints and timelines
Contribute to decision frameworks under ambiguity
Help translate marketing questions into structured, model-driven analyses
Operate in environments where experimentation is limited or infeasible, applying established methods across MMM, experiments, and observational analysis
Make and document assumptions explicitly, and flag their impact on decisions
Support decision-making under uncertainty
Provide clear analyses despite imperfect measurement, articulating trade-offs and limitations
Identify conflicting signals (e.g., attribution vs incrementality vs MMM) and discuss them with senior team members
Ensure outputs are actionable and aligned with real business constraints (budget caps, pacing, channel dependencies)
Apply methodological rigor and contribute to validation
Contribute to and apply frameworks across MMM,



incrementality testing (geo experiments, synthetic control), bidding, and/or LTV
Apply validation strategies in the absence of ground truth (cross-method validation, backtesting, sensitivity analysis)
Follow team standards for statistical rigor, interpretability, and reproducibility
Participate in knowledge sharing within the chapter and team
Communicate effectively with stakeholders
Translate modeling outputs into clear narratives for non-technical stakeholders
Communicate with squad-level stakeholders, adapting abstraction level appropriately
Address questions on model outputs with appropriate context and honesty about limitations
Build production-grade systems
Develop reliable, maintainable solutions with good standards in testing, monitoring, documentation, and reproducibility
Work closely with Engineering and Product to deploy and improve systems
Ensure long-term usability of models as decision products, not just analyses
Qualifications
Statistical, Causal & Predictive Foundations
Solid foundation in experimentation, causal inference, time-series, and predictive modeling. Experience applying these methods in real-world settings with noisy or biased data.
Working Effectively Under Ambiguity
Make sound progress on partially-defined problems, escalating when blocked
Provide reasoned analyses under uncertainty
Recognize trade-offs between conflicting methodologies and signals
Feature & Model Ownership
Track record of owning models and features from problem definition ? modeling approach ? production ? adop

📌 Data Scientist Martech (Laza)
🏢 Glovo
📍 Laza